what sets college thrivers and divers … · nber working paper series what sets college thrivers...
TRANSCRIPT
NBER WORKING PAPER SERIES
WHAT SETS COLLEGE THRIVERS AND DIVERS APART? A CONTRAST INSTUDY HABITS, ATTITUDES, AND MENTAL HEALTH
Graham BeattieJean-William P. Laliberté
Catherine Michaud-LeclercPhilip Oreopoulos
Working Paper 23588http://www.nber.org/papers/w23588
NATIONAL BUREAU OF ECONOMIC RESEARCH1050 Massachusetts Avenue
Cambridge, MA 02138July 2017
We are very grateful to Aloysius Siow, Nathalie Bau, Yosh Halberstam, Uros Petronijevic for helpful discussions. We also thank Aaron de Mello for web development and Matthew Hendrickson for helping us compile the administrative data. Financial support for this research was provided by the Ontario Human Capital Research and Innovation Fund. Any omissions or errors are our own responsibility. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research.
NBER working papers are circulated for discussion and comment purposes. They have not been peer-reviewed or been subject to the review by the NBER Board of Directors that accompanies official NBER publications.
© 2017 by Graham Beattie, Jean-William P. Laliberté, Catherine Michaud-Leclerc, and Philip Oreopoulos. All rights reserved. Short sections of text, not to exceed two paragraphs, may be quoted without explicit permission provided that full credit, including © notice, is given to the source.
What Sets College Thrivers and Divers Apart? A Contrast in Study Habits, Attitudes, andMental HealthGraham Beattie, Jean-William P. Laliberté, Catherine Michaud-Leclerc, and Philip OreopoulosNBER Working Paper No. 23588July 2017JEL No. I20,I23,I26,J20
ABSTRACT
Students from 4-year colleges often arrive having already done very well in high school, but by the end of first term, a wide dispersion of performance emerges, with an especially large lower tail. Students that do well in first year (we call the top 10 percent Thrivers) tend to continue to do well throughout the rest of their time in university. Students that do poorly (we call the bottom 10 percent Divers) greatly struggle and are at risk of not completing their degree. In this paper we use a mandatory survey with open ended questions asking students about their first-year experience. This allows us to explore more closely what sets Thrivers and Divers apart, in terms of study habits, attitudes, and personal experiences. We find that poor time management and lack of study hours are most associated with poor academic performance, and that those who struggle recognize these weaknesses. Divers also report feeling more depressed and unhappy with their lives. We posit an 'academic trap', whereby initial poor performance is related to poor time management which in turn lowers expectations, which in turn leads to lower study time, and so on. Thrivers, in contrast, study significantly more and meet with course instructors.
Graham BeattieUniversity of PittsburghDepartment of Economics4901 Wesley W. Posvar Hall230 South Bouquet StreetPittsburgh, PA [email protected]
Jean-William P. LalibertéDepartment of EconomicsUniversity of Toronto150 St. George StreetToronto, ON M5S [email protected]
Catherine Michaud-LeclercDepartment of EconomicsUniversity of Toronto150 St. George StreetToronto, ON M5S [email protected]
Philip OreopoulosDepartment of EconomicsUniversity of Toronto150 St. George StreetToronto, ON M5S 3G7CANADAand [email protected]
A online appendix is available at http://www.nber.org/data-appendix/w23588
2
I. Introduction
With postsecondary enrollments having increased dramatically over the last several decades and
the importance of degree completion continuing to be emphasized by policy makers and parents
alike, more attention is being directed towards helping post-secondary students finish and exit
with valuable experience and skills (Oreopoulos and Petronijevic, 2013). Despite efforts to
increase student support, a large fraction of students still fail to graduate. In Canada, for
example, a quarter of students registered in a four-year university program do not complete their
degree within five years (Finnie, Childs and Martinello, 2014). In the United States, the
postsecondary completion rate is less than 60 percent (NCES, 2016), and even among students
who do complete their degrees often do so with minimal grades and questionable human capital
acquisition.
Many begin university after having done very well in high school. Figure 1, for example
shows the distribution of senior-year high school grades among first year economics students at
the University of Toronto. The median acceptance grade average is 86 percent, with a standard
deviation of only 3 percentage points. However, student performance diverges widely after only
one semester. Figure 1 illustrates this increasing variance, with an especially large lower tail
from students failing or near-failing their first-semester courses. Incoming students arrive
having achieved stellar grade averages in high school, yet only a small fraction maintain these
averages into first year university, and most end up with dramatically lower grades.
Prior research has shown that previous grades are among the best predictors of how well
college and university freshman will do, but other non-academic variables also help predict
performance. Characteristics such as the Big Five personality traits – conscientiousness,
agreeableness, openness to experience, extraversion, neuroticism – and social background
sometimes rival cognitive skills in terms of predictive power for a variety of outcomes such as
educational attainment, earnings and health (Kautz et al., 2016; Almlund et al., 2011). For
instance, meta-analyses show that conscientiousness, a measure of the propensity to stay
organized, work hard and persist, is strongly positively associated with educational achievement
even after conditioning for cognitive skills (Poropat, 2009). Perseverance towards long-term
goals is also correlated with academic outcomes: the trait of grit is positively related with the
GPA of students with comparable SAT scores (Duckworth et al., 2007). Work by Mischel et al.
3
(1989) and Kirby et al. (2005) suggests that the ability to delay gratification predicts future
achievement. Many such personality constructs individually correlate with measures of academic
success. However, few data sets allow identifying which variables among all these non-academic
measures that have been found to predict college grades best explain the high variance in student
performance relative to each other.
In a previous paper (Beattie, Laliberté and Oreopoulos, 2016), we explored what
variables best predict unexpected poor or stellar grade performance in first year, after
conditioning on high school grade.1 To do so, we took advantage of a unique dataset combining a
survey of all first year economics students across all three campuses of the University of Toronto
(about 6,000 students) and administrative data containing application and performance
information. We found that while high school grade remains a strong predictor of subsequent
academic performance, several non-academic measures exhibit incremental predictive power.
Notably, conscientiousness, expected study hours and purpose-driven motivation were strongly
associated with successful and failed transition to postsecondary education.
In this paper, we provide a broader picture of the experiences students go through in first
year by following-up with them by the start of second semester. These new pieces of
information allow us to explore more closely what sets top and bottom performers apart in terms
of their study habits, attitudes, personal experiences, and health. We collect original descriptive
data that is not commonly available in order to understand better the wide dispersion in first year
performance and its correlates. We look at factors such as mental health, sense of belonging,
ambition, and study habits. We also ask students open-ended questions on their biggest struggles
in university, what they think they could change to improve, and how the university might help.
Using this unique data set, we provide a descriptive analysis of the factors that set Thrivers
(students in the top 10 percent of the grade distribution) and Divers (students in the bottom 10
percent) apart in order to understand better the main issues related to the poor performance of
first year students, which may, in turn, help suggest opportunities for additional support to
reduce poor performance.
The results show that poor time management, lack of study hours and tendencies to cram
for exams are the habits most strongly related with poor overall performance. We also find that
1 In this the current paper, we refer to students in the top decile of the end-of-semester grade distribution as Thrivers, and those in the bottom decile, Divers.
4
top performers are significantly more likely to use the free university resources available to
them and to meet with course instructors and tutors. In contrast, Divers are more likely to rely
on paid tutors. In terms of mental health, Divers are more likely to report feeling depressed and
face personal issues outside school. While a fraction of respondents believes that the university
could help by providing more mental health support and resources for non-native English
speakers, most Divers do not blame the university for their poor performance and rather suggest
that it is up to them to turn things around. However, when asked about what they should be doing
better, they often fail to come up with solutions. From this descriptive analysis, we suggest that
Divers are facing a vicious circle of poor performance, time and stress management that limit
them from identifying efficient solutions to improve their academic performance. We refer to
this vicious circle as an “academic trap”.
The remainder of the paper proceeds as follows. We describe in Section II the
randomized field experiment we implemented to collect an original set of variables on first year
students at the University of Toronto. Section III presents the results from the initial survey and
the main results from the follow up survey are discussed in Section IV. The results from the text
analysis are in Section IV. We conclude in Section V.
II. Setup and Data Collection
During the start of the 2016-17 school year, all undergraduate students enrolled in an
introduction to economics course at the three campuses of the University of Toronto were
required to complete a 'warm-up survey' and 'follow-up survey' as part of their participation
grade. Virtually all students (97 percent) completed the warm-up survey within the first three
weeks of school - a total of approximately 6,000 students. This represents about a third of all
first year students enrolled at this university, and almost 10% of the entire undergraduate student
population.2 The overall response rate for the follow-up survey was approximately 83 percent.
2 The survey was administered as part of an ongoing field experiment. Students were randomly assigned after completing the survey to different groups, with the control group receiving a one-hour personality and non-cognitive skill test, and the treatment groups receiving information on
5
The lower rate was largely due to students dropping their first year economics course, so that
there was no longer a grade incentive to complete. For students who did not drop, the response
rate was about the same as the initial survey.
The set of variables that is collected as part of the warm-up survey contains detailed
background characteristics such as international student status and parental education, as well as
a large set of self-reported measures of study habits and subjective expectations. Linked
administrative variables include gender, citizenship, registration status, GPA, all courses taken
and grades received at the University of Toronto and, for the majority of students, the high
school performance measure used for admission.3
The follow-up survey was administered at the start of the second semester (January
2017). The follow-up survey took only about 15 minutes to complete and included questions on
subjective well-being, study habits, as well as open ended questions on perceived reasons for
poor performance, the biggest challenges faced, and the type of help students could be using that
they are not currently getting.
III. Descriptive Statistics
Table 1 presents descriptive statistics from the initial survey (Fall 2016), and replicates key
results from Beattie, Laliberté and Oreopoulos (2016) using an updated cohort sample and not
conditioning on high school grade to provide a more descriptive picture.4 In column (1), we
report the mean value of the initial survey variables for all entering first year economics students.
The next three columns indicate the 25th, 50th, and 75th percentiles to provide more information
about the distribution of these variables. Column (5) shows the standardized mean difference of
these variables between Divers and the full sample (with standard errors in square brackets). For
example, self-reported weekly study hours in high school are, on average, 16 percent of a
standard deviation lower among Divers than the mean of the full sample. Column (6) shows the goal-setting, mindset, and/or study advice (see Oreopoulos and Petronijevic, 2016, for more details). 3 Admission to Canadian universities is primarily determined by academic performance in high schools. Other factors such as student essays and extracurricular activities are rarely included in application materials. 4 We do, however, condition on age, campus and cohort, to account for any differences in grading standards.
6
corresponding comparison for Thrivers. Finally, the last column indicates the mean difference
between Thrivers and Divers – which is close to 40% of a standard deviation (approximately 4.3
hours per week).
Relative to the full distribution, students who perform poorly tend to study fewer hours
both in high school and during their first year at university. They are more likely to study last
minute (0.14 s.d. above the mean) and less likely to finish what they start (0.07 s.d. below the
mean). Divers tend to work for pay more hours per week than an average student (0.16 s.d.
above the mean), they have lower expectations about their grades (0.17 s.d. below the mean) and
are less likely to aspire to go to grad school (8.5% below the mean). Unsurprisingly, they also
find the transition to college more challenging (0.11 s.d. below the mean). In terms of
demographics, international students are overrepresented in the bottom 10 percent, but first
generation status (i.e. neither of their parents is college-educated) does not predict Diver status.
On the other hand, students among the top decile are less likely to wait at the last minute
before starting studying (0.18 s.d. below the mean) and more likely to have higher expectations
with regards to their grade (0.33 s.d. above the mean). Students who excel report studying more
hours, on average, in both high school and university. They are also spending less time at a paid
job and, to some extent, checking their phone.
IV. Results
IV.A. Study Habits
Our main results come from our follow-up survey, which took place at the beginning of the
second term (January 2017), and provided an opportunity to have students reflect on their
university experiences so far.5 Recall that students had to complete both the initial and follow-up
survey to receive a 2 percent course participation grade. Virtually every student who had not
dropped the course by January completed both surveys.
Table 2 is in the same format as Table 1, comparing mean differences between Thrivers
and Divers. In Table 2, however, differences are calculated for self-reported study habits. For
5 We emphasize that the results reported in this study are aimed at mapping important associations with being a Thriver or Diver. They are not meant to imply causality between specific variables.
7
instance, one of the questions we asked students was, "Last term, when you had no immediate
deadlines (say over 2 weeks) about how many hours per week, on average, did you study
(outside of class) for all of your courses?".6 In the first row we can see that the mean response for
the full sample was 10.9 hours per week. Divers studied much less (0.30 s.d. below the mean or
approximately three hours less than the average student) and Thrivers studied over three hours
more than the average student (0.33 s.d. above the mean). This implies that on average, Thrivers
study close to seven hours per week more than Divers (p<0.01).
We also asked students how many hours they expect to study in the Winter term when
faced with no immediate deadlines. On average, students reported higher intended hours than
what they said they did in the Fall, perhaps due to unrealistic expectations (mean of 15 hours per
week). The gap in expected hours between low and high performers persists, at 38 percent of a
standard deviation. Divers, far from intending to study more than the average in the upcoming
semester to make up for their past poor performance, still expect to study significantly less than
other students.
In line with these results, Divers more often disagree with the statement that "I manage
my time well". The difference in response to this question between Divers and Thrivers is the
largest among all the study habits questions we asked (a difference of 69 percent of a standard
deviation). In addition, Divers were less likely to write down thoughts and ideas when studying
and less likely to remind themselves of their personal goals. They were less likely to meet with
their course instructor and free tutors (including TAs), but more likely to meet with an academic
advisor and paid tutors. Finally, Divers still report that they found the transition to college much
more challenging than the average student. Thrivers, on the other hand, were more likely to
report that they manage their time well and learn from their mistakes. They were more likely to
remind themselves about their goals and motivations for being in university, and they were also
more likely to meet with an instructor or free tutor. Thrivers were less likely to hire a paid tutor.
Overall, the learning habit that appears to most distinguish a Thriver from a Diver is the
ability to find sufficient time to study. Meeting with instructors and tutors, learning from past
mistakes and thinking about personal goals also sets these types of students apart.
In Figure 2, we look more carefully into the self-reported weekly study hours (excluding
hours spent in class) and show the distributions of study hours when there is no immediate
6 The appendix provides the exact wording of the survey questions.
8
deadline for students with a fall grade above 85 percent and below 65 percent. We find that a
large share of low performing students devote five hours of study per week or less. While an
important share of top students also spends less than ten hours per week, Figure 2 shows that
they are also more likely to study between 15 and 30 hours per week.
IV.B. Mental Health and Overall Experience
We now turn to the results on mental health, life satisfaction and overall university experience.
Table 3 shows that, with the exception of whether students think the university is doing its best
to help support them and whether they think their university wants them to be successful, Divers
and Thrivers significantly differ on all measures.
Students in the top decile of the grade distribution report to be more satisfied with their
life and their university experience. The gap between the average response in life satisfaction
between Thrivers and Divers is striking - more than half a standard deviation separates the two
groups. In comparison, the average subjective well-being gap between those unemployed and
working - is about 0.25 (Helliwell, 2003). Thrivers and Divers also differ markedly by how they
respond to feeling stressed and depressed.
Thrivers are more likely to report that it is important for them to do well and feel more
confident about their abilities to succeed. Since the beginning of the academic year, they have
felt less depressed or stressed. They have a higher grade average for the first term (by
construction) and much higher expectations regarding their grades for the incoming semester.
High performing students also feel that they belong at their university and, for them, being a
student at the University of Toronto is an important part of how they see themselves.
Divers are on average more likely to report a lower life satisfaction and are more likely to
have felt depressed over their first semester. Their grades during the Fall semester were more
than a full standard deviation below average, and they still expect to obtain relatively low grades
in the Winter semester (0.65 s.d. below the mean). Students in the bottom of the grade
distribution recognize that others understand the material more than themselves and are less
confident about their ability to do well, in line with prior results on the important of academic
self-efficacy, that is the belief in one’s ability to succeed (Richardson et al., 2012). We also find
9
that Divers are not as satisfied with their experience at the University of Toronto and are less
likely to feel that they belong to their university.
IV.C. Text Analysis of Student Outliers
In the follow-up survey, students were asked what their biggest challenges have been, what the
University of Toronto could do better, and what they plan to do differently. These questions
forced students to write a short response in their own words, providing a set of texts to use for a
text analysis. In this section, we focus on low performing students and present some of their
responses to understand better the main difficulties those students are experiencing.
Figure 3 lists some examples of the biggest challenges reported by the poorest
performing students (the worst 25 in the sample). Interestingly, the challenges mentioned varied
quite widely. Time management, difficulty to deal with mental health problems and personal
issues, transitioning from high school to university and language barriers are examples of
challenges divers face. Students also mention their lack of motivation to study and work. When
asked what the University of Toronto could be doing more to help, many students reported that
they felt the University was doing what it could to help and that the problem rested more on their
shoulders. But some students did point out wanting more advice on how to manage their stress
and time (see Figure 4). Better coordination on the timing of the submission of writing
assignments, especially towards the end of the semester is another example of what they think
could be improved to help them succeed better. Divers would like smaller classes and be able to
talk more with their course instructors. In Figure 5, we list some examples of what students think
they should be doing more to help themselves and succeed better. Managing time, working more
and hiring a tutor are examples of what students think they should do better. Although they see
that there is a problem and are aware what their biggest challenge is, they do not have a clear
solution. We note that some students do not know what they should be doing better.
Finally, we conduct a text analysis of the words used by Thrivers and Divers. The
methodology is identical to the text analysis in the initial survey.7 After cleaning the text through
Microsoft Word's spellcheck, we stem each word to remove grammatical endings such as verb
7 For more details about the methodology and the results from the initial survey, we refer readers to Beattie, Laliberté and Oreopoulos (2016).
10
tenses and pluralization. We then calculate a Pearson's chi-squared statistic to compare the share
of Thrivers (or Divers) who use a given word with the share of the entire sample who used it.
Very common words, such as prepositions and articles, are ignored, as are words which are used
by fewer than five Thrivers (or Divers). The results of the analysis are a set of words for each
question which are used significantly more often by Thrivers, and a second set of words which
are used significantly more often by Divers.
The results in Tables 4, 5, and 6 show several patterns. Thrivers discuss workload and
time management more often as a common concern. When asked about their biggest challenges,
they use words such as “load”, “deadline", and “hour”, and when asked about what the
University of Toronto could do help them more, they mention `”breaks”. Thrivers are also more
likely to identify these issues as things they plan to work on, as they are more likely to use words
such as “focus”, and “schedule” when asked what they will try to do differently.
This is consistent with the finding that students who end up being Thrivers are often
hard workers, as more hours spent studying allows students entering university to improve on
high school performance. It is interesting to note that Thrivers nevertheless seem to find
managing study time to be a challenge. This is consistent with the previous finding that
conscientiousness is an important factor in predicting success in the high school to university
transition, as students who can force themselves to work hard even when they do not want to are
more likely to thrive.
Further, Thrivers are more likely to identify issues such as lack of office hours,
availability of meeting time with instructors, and study space. According to our survey, they
already use these services more than Divers. This suggests that Thrivers are taking advantage of
these resources but would like more of them.
There are two major trends among Divers. First, when asked about challenges, Divers
are more likely to describe factors which may take time away from studying, using words such
as “health” and “job”. Our initial analysis identified these factors as predictors of academic
problems during the high school to university transition, but it is interesting to note that students
appear to be aware of the detrimental effects they have on their grades.
Second, Divers seem to be more lost. When asked about challenges, Divers mentioned
used words like “language” and “understand”, and when asked what the University of Toronto
could do differently, Divers mentioned they needed more information about “details”. This
11
suggests that students who did poorly transitioning to university found the experience confusing
and overwhelming.
V. Conclusion
Most students are willing to respond honestly and meaningfully to questions about life
satisfaction, well-being, and overall experiences in school. Response rates to these kinds of
questions, however, tend to be very low, making it difficult to obtain a representative sample. In
this paper, we use a survey that all first-year Economics students (6,000 students - 25 percent of
all first year students at the University of Toronto) were required to complete for participation
grade, and linked responses to administrative data. Doing so provides a rare opportunity to
uncover important differences between top and bottom performing students
The sharpest difference between Thrivers and Divers is in terms of how they spend their
time. Most Divers report studying for only 5 hours per week when not facing immediate
deadlines, and show a large tendency to cram. They also expresses awareness of their time
management problem, but nevertheless report expecting to raise study hours the following term
only slightly compared to the previous term, leaving them behind. In contrast, Thrivers study on
average more than half a standard deviation more than Divers and are much more confident
about their time management.
Thrivers are also more likely to meet with their course instructors relative to the average
student, whereas Divers tend not to know where to get advice for academic success. When asked
about what the University of Toronto could do better, some Divers suggested they would like to
have more time to talk with their professors, but most attribute internal reasons behind their
struggles.
Our results for the mental health and life satisfaction questions suggest that compared to
the average student, Divers are not enjoying their university experience. They are both more
likely to report feeling stressed and depressed. They claim that personal problems outside school
make their transition to college more challenging. The findings from the text analysis support
this result. When asked about their biggest challenges, divers were more likely to describe
factors which may take time away from studying such as “friends" and “job". Low performing
students report feeling more depressed and some believe that to perform better at school, they
12
should “try to be less depressed”. This highlights the fact that divers are less likely to know what
resources are available to them to obtain mental health support.
The main contrasts we find are consistent with what might be called an “academic trap”,
in which poor performance is related to difficulty managing time and a tendency to study last
minute. This affects and is affected by overall life satisfaction as Divers are more likely to feel
depressed and report personal issues as the main challenge to their transition to college. Because
students in the bottom of the distribution are also less likely to know where to get academic
advice and less likely to meet course instructors and free tutors, they end up lacking guidance for
identifying efficient solutions to improve their academic performance. More proactive guidance
around time management and quicker response to catch early poor performance are possible
interventions worth considering.
13
References
Almlund, Mathilde, Angela Lee Duckworth, James J. Heckman, and Tim D.
Kautz. Personality psychology and economics. No. w16822. National Bureau of Economic Research, 2011.
Beattie, Graham, Jean-William P. Laliberté, and Philip Oreopoulos. Thrivers and Divers:
Using Non-Academic Measures to Predict College Success and Failure. No. w22629. National Bureau of Economic Research, 2016.
Duckworth, Angela L., Christopher Peterson, Michael D. Matthews, and Dennis R. Kelly.
"Grit: perseverance and passion for long-term goals." Journal of personality and social
psychology 92, no. 6 (2007): 1087. Finnie, Ross, Stephen E. Childs, and Felice Martinello. "Postsecondary Student Persistence and Pathways: Evidence From the YITS-A in Canada." Research in Higher Education (2016): 1-25.
Helliwell, John F. "How's Life? Combining individual and national variables to explain subjective well-being," Economic Modelling, No. 20 (2003) pp 331-360.
Kautz, Tim, James J. Heckman, Ron Diris, Bas Ter Weel, and Lex Borghans. Fostering and
measuring skills: Improving cognitive and non-cognitive skills to promote lifetime success. No. w20749. National Bureau of Economic Research, 2014.
Kirby, Kris N., Gordon C. Winston, and Mariana Santiesteban. "Impatience and grades: Delay-discount rates correlate negatively with college GPA." Learning and individual
Differences 15, no. 3 (2005): 213-222.
Mischel, Walter, Yuichi Shoda, and Monica L. Rodriguez. "Delay of gratification in children." Science 244, no. 4907 (1989): 933.
Oreopoulos, Philip, and Uros Petronijevic. "Student coaching: How far can technology go?." Journal of Human Resources (2017): 1216-8439R.
Poropat, Arthur E. "A meta-analysis of the five-factor model of personality and academic performance." Psychological bulletin 135, no. 2 (2009): 322. Richardson, Michelle, Charles Abraham, and Rod Bond. “Psychological correlates of university students’ academic performance: a systematic review and meta-analysis.” Psychological bulletin 138, no. 2 (2012): 353
14
Figure 1: Density Distributions of Grade Averages in Grade 12 High School and Fall Term First
Year College
Note: Data for high school and fall term college grades comes from first year economics students at the
University of Toronto who participated in the initial survey of the randomized experiment.
15
Figure 2: Density Distributions of Weekly Average Study Hours According to Fall Grade
Note: Data for self-reported study hours comes from the follow-up survey of the field experiment.
16
Figure 3: Biggest Challenge Among Lowest Performing Students
Sample Phrases
Trying to manage time efficiently
Not being able to focus properly and not getting the mental health treatment as regularly as I would like.
I'm still in the transaction from high school to university and I felt stressed very much. I'm also experiencing lots of stress from my family and relationship.
Adjusting to how different my 3rd world country high school was in comparison to UofT
Having no friends and the fact that no one wants to talk to me in my classes.
Not enough inspiration for me to myself and sometimes I don't know which career should I have.
For me, the biggest challenge is language, because English is not my first language, so I have to take more time to understand what prof. teaches.
Finding motivation to actually sit down and study, in high school I got As without having to do any work or study.
depression has been a huge hindrance in finding motivation to do work/go to class
Sometimes I feel like I am ready for the test or exam. However, after the exam I feel like I don't know anything from my review.
Note: The sample phrases are taken from the responses of divers to the open question “What have been the biggest challenges to your academic success so far?” in the follow-up survey.
17
Figure 4: What could the University of Toronto be doing more to help?
Sample Phrases
I think UofT is doing the best it could, now it's up to me.
I think UofT should create more groups to help those international students in order to help them with their language problems.
Coordinate the timing of assignments and test so that they are better interspersed especially towards the end of the semester.
Give students more advice about time management and how to manage stress.
Make the class size smaller. More time talking to professors.
Note: The sample phrases are taken from the responses of divers to the open question “What could UofT be doing more to help?” in the follow-up survey.
Figure 5: What could YOU be doing more to help?
Sample Phrases
try to be less depressed
Focusing a lot more in class and doing all my work.
I am thinking about how I will take a more diverse approach in seeking self-enhancement, instead of mere academic success.
I need to have a tutor to help me study.
Managing time better
No idea to be honest
Note: The sample phrases are taken from the responses of divers to the open question “What could you be doing differently?” in the follow-up survey.
Tab
le1:
Init
ial
Surv
eyR
esp
onse
sA
mon
gSurv
eySam
ple
ofE
conom
ics
Stu
den
tsG
roup
edby
Bot
tom
and
Top
Dec
ile
inE
nd-o
f-Sem
este
rG
rades
Bo
tto
m d
ecile
Top
dec
ile
Dif
fere
nce
bet
wee
n o
utl
iers
Mea
n2
5th
50
th7
5th
Mea
n d
iff.
Mea
n d
iff.
(6)
- (5
)
[s.d
.]p
erce
nti
lep
erce
nti
lep
erce
nti
le[s
.e.]
[s.e
.][p
-val
tes
t (5
)=(6
)]
Surv
ey v
ari
ab
les
(1)
(2)
(3)
(4)
(5)
(6)
(7)
Stu
dy
ho
urs
in H
igh
Sch
oo
l1
2.4
49
18
-0.1
63
***
0.2
25
***
0.3
89
[11
.05
0]
[0.0
43
][0
.04
3]
[0.0
00
]
Stu
dy
ho
urs
th
is y
ear
17
81
52
4-0
.14
6**
*0
.21
8**
*0
.36
3
[11
.81
0]
[0.0
43
][0
.04
3]
[0.0
00
]
Thin
k ab
ou
t fu
ture
go
als
2.3
84
12
30
.05
1-0
.08
4*
-0.1
35
[1.2
12
][0
.04
3]
[0.0
43
][0
.02
8]
Last
min
ute
stu
dy
4.0
77
34
50
.14
3**
*-0
.17
8**
*-0
.32
1
[1.5
09
][0
.04
3]
[0.0
43
][0
.00
0]
Tim
es c
hec
ked
ph
on
e la
st h
ou
r2
.72
41
35
-0.0
56
-0.0
73
*-0
.01
7
[1.7
34
][0
.04
3]
[0.0
43
][0
.77
8]
Oft
en c
hec
k p
ho
ne
at s
cho
ol
0.4
42
00
1-0
.01
0-0
.04
2*
-0.0
32
[0.4
97
][0
.02
1]
[0.0
22
][0
.29
7]
Wo
rk h
ou
rs p
er w
eek
6.5
69
03
10
0.1
62
***
-0.1
32
***
-0.2
94
[8.8
85
][0
.04
3]
[0.0
43
][0
.00
0]
Exp
ecte
d g
rad
e8
3.2
27
88
28
7-0
.17
2**
*0
.33
1**
*0
.50
3
[6.6
32
][0
.04
3]
[0.0
43
][0
.00
0]
Tran
siti
on
to
co
llege
3.2
39
23
4-0
.10
7**
0.0
77
*0
.18
4
[1.5
98
][0
.04
3]
[0.0
43
][0
.00
3]
I fin
ish
wh
at I
star
t3
.83
33
44
-0.0
72
*0
.02
40
.09
6
[0.8
02
][0
.04
3]
[0.0
43
][0
.11
8]
Setb
acks
do
n't
dis
cou
rage
3.4
16
33
4-0
.00
7-0
.01
7-0
.01
[0.9
67
][0
.04
3]
[0.0
43
][0
.87
5]
Wan
t m
ore
th
an B
A0
.67
01
1-0
.08
5**
*0
.03
7*
0.1
22
[0.4
70
][0
.02
0]
[0.0
20
][0
.00
0]
Firs
t ge
ner
atio
n0
.29
70
01
0.0
30
-0.0
49
**-0
.08
[0.4
57
][0
.02
0]
[0.0
20
][0
.00
4]
Firs
t ye
ar s
tud
ent
0.8
02
11
1-0
.03
5**
-0.0
90
***
-0.0
55
[0.3
98
][0
.01
7]
[0.0
17
][0
.02
4]
Inte
rnat
ion
al s
tud
ent
0.4
63
00
10
.08
5**
*0
.01
3-0
.07
2
[0.4
99
][0
.02
1]
[0.0
22
][0
.01
8]
Raw
var
iab
les
Stan
dar
diz
ed v
aria
ble
s
Note
s: R
esu
lts
are
bas
ed o
n 5
,35
6 s
tud
ents
. In
colu
mn
(1
), t
he
mea
n a
nd
sta
nd
ard
dev
iati
on
of
each
var
iab
le i
s re
port
ed. C
olu
mn
s (2
) th
rou
gh
(4
) sh
ow
th
e
25
th, 5
0th
an
d 7
5th
per
cen
tile
s. I
n c
olu
mn
s (5
) an
d (
6),
th
e co
effi
cien
ts r
epre
sen
t th
e d
iffe
ren
ce i
n m
ean
s, i
n s
tan
dar
d d
evia
tion
un
its
for
con
tin
uou
s var
iab
les
and
in
per
cen
tage
poin
ts f
or
bin
ary v
aria
ble
s, b
etw
een
ou
tlie
r gro
up
s an
d t
he
full
sam
ple
. C
olu
mn
(7
) d
isp
lays
the
dif
fere
nce
bet
wee
n o
utl
iers
an
d r
eport
th
e
p-v
alu
e of
a te
st o
f m
ean
dif
fere
nce
. *** p
<0
.01
, ** p
<0
.05
, * p
<0
.1
Tab
le2:
Sel
f-R
epor
ted
Stu
dy
Hab
its
(sta
rtof
seco
nd
term
)A
mon
gSurv
eySam
ple
ofE
conom
ics
Stu
den
tsG
roup
edby
Bot
tom
and
Top
Dec
ile
inE
nd-o
f-Sem
este
rG
rades
Bo
tto
m d
ecile
Top
dec
ile
Dif
fere
nce
bet
wee
n o
utl
iers
Me
an2
5th
50
th7
5th
Me
an d
iff.
Me
an d
iff.
(6)
- (5
)
[s.d
.]p
erce
nti
lep
erce
nti
lep
erce
nti
le[s
.e.]
[s.e
.][p
-val
tes
t (5
)=(6
)]
Follo
w-u
p s
urv
ey
vari
able
s(1
)(2
)(3
)(4
)(5
)(6
)(7
)
Stu
dy
ho
urs
th
is F
all
10
.91
47
15
-0.3
01
***
0.3
30
***
0.6
31
[10
.50
0]
[0.0
57
][0
.05
0]
[0.0
00
]
Stu
dy
ho
urs
th
is t
erm
14
.88
61
02
0-0
.15
9**
*0
.22
0**
*0
.37
9
[12
.34
0]
[0.0
58
][0
.05
0]
[0.0
00
]
Man
age
my
tim
e w
ell
3.6
62
34
4-0
.38
5**
*0
.30
5**
*0
.69
[1.1
79
][0
.05
7]
[0.0
50
][0
.00
0]
Lear
n f
rom
my
mis
take
s4
.65
24
55
-0.3
34
***
0.2
72
***
0.6
06
[1.0
14
][0
.05
7]
[0.0
50
][0
.00
0]
Wri
te d
ow
n id
eas
wh
en s
tud
yin
g4
.11
93
45
-0.1
67
***
0.1
79
***
0.3
46
[1.2
30
][0
.05
8]
[0.0
50
][0
.00
0]
Get
fee
db
ack
for
wri
tin
g as
sigm
ents
3.6
83
45
-0.0
83
0.1
01
**0
.18
4
[1.2
68
][0
.05
8]
[0.0
50
][0
.01
6]
Rem
ind
mys
elf
of
my
goal
s4
.47
74
55
-0.1
76
***
0.1
32
***
0.3
07
[1.1
42
][0
.05
8]
[0.0
50
][0
.00
0]
Tran
siti
on
has
bee
n c
hal
len
gin
g4
.34
24
45
0.1
80
***
-0.2
25
***
-0.4
05
[1.2
53
][0
.05
8]
[0.0
50
][0
.00
0]
Me
t w
ith
Inst
ruct
or
1.3
61
01
2-0
.15
8**
*0
.41
4**
*0
.57
2[1
.82
2]
[0.0
57
][0
.05
0]
[0.0
00
]
Me
t w
ith
Ad
viso
r0
.89
40
01
0.1
20
**-0
.04
0-0
.16
[1.3
36
][0
.05
8]
[0.0
50
][0
.03
6]
Me
t w
ith
Fre
e tu
tor
1.6
90
12
-0.1
96
***
0.2
28
***
0.4
23
[2.2
26
][0
.05
8]
[0.0
50
][0
.00
0]
Me
t w
ith
Pai
d t
uto
r1
.01
00
10
.11
2*
-0.1
48
***
-0.2
6
[2.0
78
][0
.05
8]
[0.0
50
][0
.00
1]
We
nt
to w
ork
sho
ps
1.1
88
00
2-0
.06
2-0
.01
90
.04
3
[1.8
24
][0
.05
8]
[0.0
50
][0
.57
1]
Par
tici
pat
e in
ext
ra c
urr
icu
lar
0.3
46
00
1-0
.11
5**
*0
.05
5**
0.1
69
[0.4
76
][0
.02
7]
[0.0
24
][0
.00
0]
Raw
var
iab
les
Stan
dar
diz
ed v
aria
ble
s
No
tes:
Res
ult
s ar
e b
ased
on
3,8
49
stu
den
ts. I
n c
olu
mn
(1
), t
he
mea
n a
nd
sta
nd
ard
dev
iati
on
of
each
var
iab
le i
s re
po
rted
. C
olu
mn
s (2
) th
rou
gh
(4
) sh
ow
th
e
25
th, 5
0th
an
d 7
5th
per
cen
tile
s. I
n c
olu
mn
s (5
) an
d (
6),
th
e co
effi
cien
ts r
epre
sen
t th
e d
iffe
ren
ce i
n m
ean
s, i
n s
tan
dar
d d
evia
tio
n u
nit
s fo
r co
nti
nu
ou
s var
iab
les
and
in
per
cen
tage
po
ints
fo
r b
inar
y v
aria
ble
s, b
etw
een
ou
tlie
r gro
up
s an
d t
he
full
sam
ple
. C
olu
mn
(7
) d
isp
lays
the
dif
fere
nce
bet
wee
n o
utl
iers
an
d r
epo
rt t
he
p-v
alu
e o
f a
test
of
mea
n d
iffe
ren
ce. *** p
<0
.01
, ** p
<0
.05
, * p
<0
.1
Tab
le3:
Fol
low
-up
Res
pon
ses
(sta
rtof
seco
nd
term
)A
mon
gSurv
eySam
ple
ofE
conom
ics
Stu
den
tsG
roup
edby
Bot
tom
and
Top
Dec
ile
inE
nd-o
f-Sem
este
rG
rades
Bo
tto
m d
ecile
Top
dec
ile
Dif
fere
nce
bet
wee
n o
utl
iers
Me
an2
5th
50
th7
5th
Me
an d
iff.
Me
an d
iff.
(6)
- (5
)
[s.d
.]p
erce
nti
lep
erce
nti
lep
erce
nti
le[s
.e.]
[s.e
.][p
-val
tes
t (5
)=(6
)]
Follo
w-u
p s
urv
ey v
aria
ble
s(1
)(2
)(3
)(4
)(5
)(6
)(7
)
Life
sat
isfa
ctio
n4
.50
64
55
-0.3
66
***
0.2
22
***
0.5
88
[1.3
65
][0
.05
7]
[0.0
50
][0
.00
0]
Sati
sfac
tio
n w
ith
Uo
fT4
.39
64
45
-0.3
13
***
0.3
10
***
0.6
23
[1.3
22
][0
.05
7]
[0.0
50
][0
.00
0]
Gra
de
this
Fal
l7
3.5
68
75
78
-1.1
19
***
0.9
97
***
2.1
16
[9.5
47
][0
.05
2]
[0.0
45
][0
.00
0]
Exp
ecte
d g
rad
e th
is t
erm
80
.33
75
82
87
-0.6
50
***
0.5
94
***
1.2
44
[7.7
99
][0
.05
6]
[0.0
48
][0
.00
0]
Bel
on
g at
Uo
fT4
.19
74
45
-0.1
81
***
0.1
68
***
0.3
49
[1.0
99
][0
.05
8]
[0.0
50
][0
.00
0]
Iden
tify
wit
h U
ofT
4.3
62
45
5-0
.10
3*
0.0
68
0.1
71
[1.2
02
][0
.05
8]
[0.0
50
][0
.02
6]
Uo
fT w
ants
me
to s
ucc
eed
4.0
84
34
50
.02
10
.13
6**
*0
.11
5
[1.2
83
][0
.05
8]
[0.0
50
][0
.13
3]
Kn
ow
wh
ere
to g
et
advi
ce4
.30
84
45
-0.1
40
**0
.13
4**
*0
.27
5
[1.1
63
][0
.05
8]
[0.0
50
][0
.00
0]
Uo
fT s
up
po
rts
me
4.0
21
34
5-0
.10
3*
0.1
32
***
0.2
35
[1.1
40
][0
.05
8]
[0.0
50
][0
.00
2]
Oth
ers
un
der
stan
d m
ore
th
an m
e3
.93
34
50
.23
8**
*-0
.03
7-0
.27
4
[1.1
57
][0
.05
8]
[0.0
50
][0
.00
0]
Imp
ort
ant
for
me
to d
o w
ell
5.9
24
56
7-0
.20
0**
*0
.16
7**
*0
.36
7
[1.1
95
][0
.05
8]
[0.0
50
][0
.00
0]
Co
nfi
den
t to
do
wel
l4
.34
83
45
-0.3
88
***
0.3
18
***
0.7
05
[1.4
89
][0
.05
7]
[0.0
50
][0
.00
0]
Dep
ress
ed
1.5
87
12
20
.22
7**
*-0
.15
2**
*-0
.37
9
[0.8
69
][0
.05
8]
[0.0
50
][0
.00
0]
Stre
sse
d1
.93
51
23
0.1
59
***
-0.1
74
***
-0.3
33
[0.8
34
][0
.05
8]
[0.0
50
][0
.00
0]
Raw
var
iab
les
Stan
dar
diz
ed v
aria
ble
s
No
tes:
Res
ult
s ar
e b
ased
on
3,8
49
stu
den
ts. In
co
lum
n (
1),
th
e m
ean
an
d s
tan
dar
d d
evia
tio
n o
f ea
ch v
aria
ble
is
rep
ort
ed. C
olu
mn
s (2
) th
rou
gh
(4
) sh
ow
th
e
25
th, 5
0th
an
d 7
5th
per
cen
tile
s. I
n c
olu
mn
s (5
) an
d (
6),
th
e co
effi
cien
ts r
epre
sen
t th
e d
iffe
ren
ce i
n m
ean
s, i
n s
tan
dar
d d
evia
tio
n u
nit
s, b
etw
een
ou
tlie
r gro
up
s
and
th
e fu
ll s
amp
le. C
olu
mn
(7
) d
isp
lays
the
dif
fere
nce
bet
wee
n o
utl
iers
an
d r
epo
rt t
he
p-v
alu
e o
f a
test
of
mea
n d
iffe
ren
ce. *** p
<0
.01
, ** p
<0
.05
, * p
<0
.1
Table 4: Biggest Challenges
What have been the biggest challenges to your academic success so far?
Top decile words Bottom decile words
Significant
at 99%
office first
Significant
at 95%
quiz commute load day deadline
anziety mark
job language issue understand
Significant
at 90%
write speak track science hour health fail
Sample Phrases
office TA office hours have helped me a lot.
Need more office hours
I don’t go to office hours or tutorials even though I should.
quiz Keeping calm during quiz (not test).
the constant eco100 quizzes are a pain and ruin my sundays.
quiz/exams just keep coming
commute imposing a 2-hour commute each way via public transit
I have a two hour commute to school.
Having to commute to classes
load learning how to manage the course load has been quite difficult.
study load and work off campus
The intense teaching pace as well as heavy work load
day Waking up late causes me to have less time to study throughout the day
some days where I would procrastinate and would not get any work done
i spent 5 days to study
deadline Having the motivation to study way before deadlines.
Difficulty in meeting deadlines
getting things done at an appropriate time before deadlines.
anxiety Coping with with stress and anxiety.
I burn out easily and experience anxiety and panic attacks.
The anxiety that I face personally has held me back
mark Dealing with the pretentious and ambiguous marking scheme.
Ridiculously difficult tests and extremely hard marking
the marking is much more subjective and vague than it should be.
write Language difficulties including writing and speaking.
Theres not much more I can write unfortunately.
Writing assignments and presentations.
speak speaking in a large group of people making me stressed out.
To achieve a higher grade (over 90) generally speaking.
For example, speaking and writing.
track I was able to do enough to stay on track.
stay focused and on track with my studying.
Adjusting my time in able to keep everything on track.
science In my science courses, the amount of material is massive
language and computer science,the boring life in the residence
CSC148 and CSC165 so I can get into the computer science program.
hour Commuting and long hours
I always wish there’s 48 hours a day.
3 hour lectures are boring
first Transitioning from high school to university first year
The teacher doesn’t use English since the first class.
I have not had any real challenges since first year
job I have an partime job
Balancing my time between my part time job, my lectures and studying.
I did a very great job at my academic life
language Meeting with new people and making friends due to language barriers.
I don’t know being away from family, speaking different languages,
biggest challenge is language, because English is not my first language,
issue I haven’t acknowledged the issue until recently
Time management has been a really big issue.
Family issues triggered how I did and depression.
understand I can not understand what prof said in class from time to time
making sure I understand the material well
I don’t understand most of the course
health not getting the mental health treatment as regularly as I would like.
the plague it has been to my life and mental health.
because I faced a lot of problems with my health
fail i just dont understand how i failed it.
I tried to pass the MGM101 but failed,
I have failed to do what I should have done
Table 5: What could the University of Toronto be doing more to help?
What could U of T be doing more to help?
Top decile words Bottom decile words
Significant
at 99%
lack run
Significant
at 95%
meet math space material friend kind reach
Significant
at 90%
hour break advice instructor
course
international detail
Sample Phrases
lack I was refused by the department because of the lack of prerequisite.
But I feel that there is a lack of TA’s,
Nothing really, it’s just a lack of motivation .
run school events that run late into the night should be on fridays
instead of running back and forth between family doctors and school,
networking event that runs far into the evening.
meet I rarely get to meet my profs one-on-one to ask for help
clear how to meet with an academic advisor and what that entails.
I sometimes feel not comfortable being at a face-to-face meeting
math Maybe more math aids for higher level courses?
Wednesday is a popular day for a lot of math and csc courses.
The math aid centre is understaffed,
space More comfortable study spaces
UofT could provide more studying spaces,
The UTSC campus could also use more study spaces.
material the difficulty of the material is reasonable.
practice questions and tests that align with the material covered in tests.
inconvinent to search and use academic material outside the campus
hour make some more ta office hours
More office hours and less homework.
Try to have less 3-hour lecture blocks
break fall reading week instead of the short two day fall break.
Give longer winter break.
More break would be better.
advice Give first year students more advice on how to proceed.
mentorship and advice from people who went through the SAME thing
career advice, offer internship
instructor There are only a few instructors, so it is hard to ask questions.
sometimes instructors have to spend more time on erasing the boards.
To make the information about instructors and tutors more accessible on
websites.
course help students with their work and understanding of course materials
Offer more online courses.
Provide more chances for students who take the same course
friend I can study with my friends in the study room.
my friend was kicked out just because he did not pass the ace language.
more event that student can attend to meet new friend
kind have more different kinds of activities
helping students find the right kind of help.
Nothing, it’s kind of a personal issue.
reach Have more people to reach out to when i have a problem
Help is out there, all I need to do is to reach for it.
If it could reach out or promote its program more like its health,
international setting a group for helping international students.
course for international student to teach them how to use all the resources
workshop for international student to learn how to read effectively.
detail explain more details when first years students are choosing courses.
give more details about the test.
need more detailed help since we have few people than other campuses.
Table 6: What could YOU be doing more to help?
Thinking of trying differently
Top decile words Bottom decile words
Significant
at 99%
break mean follow campus place
class person
Significant
at 95%
day instructor focus question ask
helpful schedule
efficient
Significant
at 90%
earlier read term idea
Sample Phrases
break I will go to sleep/take a nap/do some yoga/take a walk/take a break.
efficiently take breaks between studying.
I will find myself a volunteer at pharmacy during summer break
mean But by all means the best help will always be financial
what does it mean?
I’m not quite sure what this question means,
follow stay focused/consistent by following the schedule
following a document on how to identify and correct negative thoughts
set clear goals and follow them through
campus Living off campus is really a disadvantage for commuters.
help me excel in this pathetic campus.
be more participated with the campus about everything.
place Studying in the library, or any places without distractions.
Trying to find a place near campus to live.
choosing a place with less distractions
class studying an hour a day for each class i have a day
Going to class more and take more in depth note, and paying attention.
when I have questions during the class, I will ask professor
person Put myself out there and talk to people.
After all, we do have a rather large number of people here.
join clubs/groups with people who are interested in similar things as me.
day Use a planner and try to plan out my day before it starts
I could make a to do list every day, plan my day in advance.
I need to plan for each day and week.
instructor Having more meetings with the course instructors
if I turn to instructors for help
meet with instructors more often
focus I’m focusing on working harder and getting less distracted.
it will help me stay focused and relaxed.
focus more on the necessary ones
question I’m not sure about what this question is asking for.
Don’t want to answer this question. Why I cannot leave it blank :(
I don’t understand the question English is my second language
ask I will be less hesitant to approach people and to ask them questions.
ask for friends’ help.
Study more in quiet spaces and ask for help when I have questions.
helpful trying something differently is always helpful
Of course it would be much more helpful than before.
However, the service is not helpful.
schedule I have visited an academic advisor, and plan on having a good schedule
I can get more time off and have a more stress free schedule/timetable
I am trying to have a more concrete schedule of what to study
earlier Get things done earlier rather than putting them off.
Starting to study earlier than before.
Maybe trying to choose courses earlier.
read Spending weekdays doing course work (e.g., readings)
Do my readings before class.
prioritizing my schedule and focusing on readings when I am reading
term I do need to reschedule in this term,
Studying more and working out once a week this term
seeking help from writing centre instructors this term
idea Have no idea.
i don’t have a idea
Sounds great, but I have no specific idea.
efficient improve my efficient of study.
I should work more efficient.
trying to manage my time more efficiently.